# PaddlePaddle/VisualDL

Deep Learning Visualization Toolkit（『飞桨』深度学习可视化工具 ）

Repository: https://github.com/PaddlePaddle/VisualDL
Canonical: https://ross.abutalabs.com/products/visualdl
Homepage: https://www.paddlepaddle.org.cn/paddle/visualdl
Language: HTML
License: Apache-2.0
License Family: permissive
Topics: visualization, deep-learning, paddlepaddle, onnx, caffe
Last push: 2025-01-22T06:02:52+00:00

## Health v2 (maintenance only)
Score: 24/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 2, release rhythm 8, longevity 100
- inputs: {"age_days": 3178, "days_push": 588, "days_rel": 672, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 4884, forks 631 (observed 2026-08-28T04:09:01.845935+00:00)

## What it is
VisualDL is a deep learning visualization toolkit for PaddlePaddle that provides charts for tracking training metrics, visualizing model structures, tensor histograms, PR/ROC curves, and high-dimensional data embeddings. It is a Python-based tool with a browser UI for understanding and optimizing model training.

## Use cases
- visualize training loss and accuracy curves in real time
- inspect a deep learning model's computation graph
- plot PR and ROC curves for classification experiments
- project high-dimensional embeddings to 2D/3D
- compare hyperparameters against model metrics
- share experiment results with a team

## When to choose
- you train models with PaddlePaddle and want built-in visualization
- you need a TensorBoard-like tool with ONNX model graph support
- you want lightweight Python integration with a few lines of code

## When to avoid
- you need a full experiment management platform with team collaboration and model registry
- your stack is PyTorch/TensorFlow and you already use TensorBoard or Weights & Biases

## Facets
- artifact type: application
- maturity: stable
- function: data-visualization, monitoring, machine-learning, deep-learning
- domain: deep-learning, data-visualization, machine-learning, developer-tools
- platform: python, cross-platform
- tags: experiment-tracking, tensorboard-alternative, paddlepaddle, onnx, model-visualization, training-metrics, web-server

## Member repositories
- PaddlePaddle/VisualDL (main) score 24

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:01.845935+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:18:14.491692+00:00, confidence not recorded.
  - readme: https://github.com/PaddlePaddle/VisualDL (fetched 2026-08-28T04:09:01.845935+00:00, sha ab21144784e3)
  - homepage: https://www.paddlepaddle.org.cn/paddle/visualdl (fetched 2026-08-29T09:00:36.418884+00:00, sha 6c227ac7c0f8)
- Data as of 2026-08-30T08:39:29.467469+00:00.